RE: Libpq support to connect to standby server as priority

From: "Tsunakawa, Takayuki" <tsunakawa(dot)takay(at)jp(dot)fujitsu(dot)com>
To: 'Haribabu Kommi' <kommi(dot)haribabu(at)gmail(dot)com>
Cc: Robert Haas <robertmhaas(at)gmail(dot)com>, Laurenz Albe <laurenz(dot)albe(at)cybertec(dot)at>, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>, Dave Cramer <pg(at)fastcrypt(dot)com>, Jing Wang <jingwangian(at)gmail(dot)com>, PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org>
Subject: RE: Libpq support to connect to standby server as priority
Date: 2019-03-29 08:05:46
Message-ID: 0A3221C70F24FB45833433255569204D1FBED011@G01JPEXMBYT05
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Hi Hari-san,

I've reviewed all the files. The patch would be OK when the following have been fixed, except for the complexity of fe-connect.c (which probably cannot be improved.)

Unfortunately, I'll be absent next week. The earliest date I can do the test will be April 8 or 9. I hope someone could take care of this patch...

(1) 0001
With this read-only option type, application can connect to
to a read-only server in the list of hosts, in case
...
before issuing the SHOW transaction_readonly to find out whether

"to" appears twice in a row.
transaction_readonly -> transaction_read_only

(2) 0001
+ succesful connection or failure.

succesful -> successful

(3) 0008
to conenct to a standby server with a faster check instead of

conenct -> connect

(4) 0008
Logically, recovery exit should be notified after the following statement:

XLogCtl->SharedRecoveryInProgress = false;

(5) 0008
+ /* Update in_recovery status. */
+ if (LocalRecoveryInProgress)
+ SetConfigOption("in_recovery",
+ "on",
+ PGC_INTERNAL, PGC_S_OVERRIDE);
+

This SetConfigOption() is called for every RecoveryInProgress() call on the standby. It may cause undesirable overhead. How about just calling SetConfigOption() once in InitPostgres() to set the initial value for in_recovery? InitPostgres() and its subsidiary functions call SetConfigOption() likewise.

(6) 0008
async.c is for LISTEN/UNLISTEN/NOTIFY. How about adding the new functions in postgres.c like RecoveryConflictInterrupt()?

(7) 0008
+ if (pid != 0)
+ {
+ (void) SendProcSignal(pid, reason, procvxid.backendId);
+ }

The braces are not necessary because the block only contains a single statement.

Regards
Takayuki Tsunakawa

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Hadi Moshayedi 2019-03-29 08:06:31 Re: Multitenancy optimization
Previous Message Michael Paquier 2019-03-29 08:04:26 Re: REINDEX CONCURRENTLY 2.0